Do you use/code Algos? Is technical analysis a meme?
Algorithms
Landon Lee
Other urls found in this thread:
Lucas Robinson
If you want a completely automated, 100% trading system, guaranteed to work in every single market context/environment/liquidity situation/fundamental situation, and which "predicts the future" - then I'm afraid TA is a big meme.
Have fun competing over arbitrage opportunities, whether statistical, or otherwise. Enjoy paying shitloads for past order data, to find market microstructure inefficiencies that MIGHT exist, in certain situations, and which have somehow not been discovered by 100 funds before you. Have fun figuring out all the complicated order types exchanges can offer, and competing with the HFTs over shit like queue position.
TA gives so much, for so little. Of course it's not going to be perfect, it's just a fucking tool. Get over it.
Alexander Robinson
>TA gives so much, for so little.
Typo? (No Sarcasm)
>Funds know what you know.
Why can't you take a few crumbs of the pie?
You're a pure fundamental analyst then?
Ayden Carter
No, I use technical analysis.
And if you ask shit like "Does TA work?", I'm going to be honest and say, you're not going to make it. You need to take initiative, do your own forms of testing, and come to your own conclusions. Not listen to other peoples' stupid fucking opinions - whether they're for a certain approach, or not.
I don't have enough knowledge to comment on your second question. But I'd suspect it to be unlikely.
Levi Brooks
>do your own forms of testing
Isn't that where coding an algo is useful
John Baker
It will depend on the sort of strategy you want to focus on, and the market(s) you'll focus on.
If want to do stuff that involves spread trades between instruments, pairs trading, or statistical arbitrage - backtesting those sorts of strategies are an absolute must, BEFORE you actually deploy a strategy.
But, if you're going to do something more TA based, I'd recommend firstly understanding how the market you want to trade, actually "works". Stuff like how does it move during different timezone sessions, during trend days, during range days, during range expansion days, during moments before and after certain types of news. Take records of this, and make sure to look at how things work on multiple time frames, not just one.
Then, begin to work towards the strategy/ies you'll use, and how you'll identify the precursors of the market conditions those strategies/y would work best in. It takes a lot of work to do that. But if you want to truly understand how markets will trade in the technical setups you'll eventually backtest, then that's the approach I'd recommend.
People don't put the effort in to do this, they want a quick solution to testing. It doesn't work like that.
Zachary Sullivan
Saved this. Thank you.
Logan White
I found an actual strategy that's better than MA or levels, and it's pretty simple if you think about it. Why the most (worse) indicators are pushed so hard, is because they don't want you to think for yourself and actually discover winning strategies.
I'm not a quant or anything, but it's kind of obvious that nobody is going to tell you any method or strategy that they are actually making money with, for the fear of crowding out the trade.
Most TA gurus like Tone Vays are just selling you lessons, affiliate marketing and "education" because they don't actually make any money trading.
Jason Parker
>Most TA gurus like Tone Vays are just selling you lessons, affiliate marketing and "education" because they don't actually make any money trading.
proof?
Adrian Garcia
See
>Tone Vays' short at 6100
Isaac Hill
If you have a winning strategy why make it public so everyone can use it and it doesn't work anymore? If you're rich off this strategy why waste time making nothing being a teacher?
John Kelly
>pure speculation and cope
Just as expected.
Caleb Price
Asked you a question. And coping with what?
Aiden Edwards
>Asked you a question
>he thinks he can command random people
>he has no reading comprehension
Ah, the typical brainlet who can't cope with people doing better then him.
William Peterson
Even shallow machine learning techniques should easily find any TA-type pattern in the data. Problem starts if some of the trading agents have strategies that depend on their internal state - then system becomes nonergodinc and unapproachable by statistical learning, but according to coworker that trades with bots simple pattern following strategies are still viable for now...
James Young
youre trying too hard ;)
Joseph Sanchez
>shallow machine learning techniques should easily find any TA-type pattern in the data
You or Co-worker has done this?
Henry Baker
To add details, "the strategy" that supposedly works for him is looking at about 20 markets for certain pattern that appears couple times a day and then executing trades that (hopefully) make about 1-2% profit. Just candles, probably no order book, definitely no external data (sentiment on twitter anyone?). Custom, simple and crude bot running on a spare PC...
Blake Kelly
Machine learning isn't magic. It doesn't do research for you either.
Caleb Davis
CSW here. Let me give you brainletses a pro tip. A good way to make money is by selling something to a lot of people, for a lot of money, at a low cost. Obviously you traders get this already, you can skip forward to appendage A.
appandage A: What I did when I invented Bit Coins was tap into an existing infrastructure. Half a billion street shitting millennials who couldn't wait any longer for the chink to collapse. The same number in chinks themselves that missed out, and a billion other poor, retarded, absolute faggots around the world.
These are the sub humans working, for free, on my masterpiece. They compete among themselves to fork existing scams and schemes into wealth transfer mechanisms of their own, they lobby muh gubberment, they want muh gibs.
Meme liners sit toward the bottom of this barrel. They believe by doing literally nothing they could build a wealth transfer pipeline directly to their own pocket, a pipeline so big they could be rich. They aren't bad people, they're just too retarded to get it.
Carson Walker
Oh absolutely. Looking if whitepaper makes any sens or if the team can execute/communicate is imho essential for investing. But on the other hand I can understand why people don't bother with those for "just trading", especially if it's to be done by the bot and you don't care about anything that happens after.
Not personally, but I've had some experience with ml in the past and see no reason why any as glaringly obvious pattern as commonly discussed TA patterns wouldn't be found by ml.
I also expect (note that this is separate claim from above!) most attempts go like this youtube.com
Jayden Davis
So, I've done this before, many times, but I'll show you guys how I use TA, mine being based off of Mutiple Time Frames, especially the 200SMA.
My trading style is extremely discretionary, this is not how everyone uses TA, and so, I don't want to group all other TA traders with me, because they'll have their own approaches.
I'll show you a trade I did a while back. I don't believe markets can be predicted, only reacted to.
This will show you how much of a dirty Battlefield the market actually is. And fights take place around Objective Technical points, like Moving Averages. If you make regular observations, you'll realize this. It's not something mathematical, or secret.
1. I first generate my technical thesis off of a higher time frame. Here it was a Fib retrace, there had been some Chinese trade news driving the Dow down.
Cameron Ward
2. first entry made, and you'll see that I don't end up following my second entry plan, which ends up going against me.
Justin Perry
3. Second entry made at Resistance, rather than where I had initially planned.
Jeremiah Hill
4. Resistance is held, and I start to shit myself.
Anthony Perez
5. Realize it occurred due to the 30min 5SMA being held.
Lincoln Bennett
8. I start considering exits.
Xavier Garcia
9. Support is held, but now I'm scared, because my sizing is off, because I didn't follow my plan.
Gavin Ward
10. Bears are still trying to press down, using whatever moving average they can find.
Jayden Cruz
11. I still stick with the trade because of Support justifications.
Justin Collins
12. And now, watch how a massive fight takes place at the 200SMA, it is an extremely important moving average, all of us know its importance.
Alexander Young
13. They pressed down again, but I still held on to the trade, due to the Support shown. And now progress is being made.
Ryder Adams
I have a PhD in data science and can say with absolute certainty that you're a fucking clown. This is called 'story telling' in the finance industry.
Bentley Martinez
14. Then boom. Bulls manage to press hard enough to start hitting some stops, those Bears in profit start exiting, and you get this big movement up as the 200SMA is cleared.
Noah Brooks
This. or 'autism' where i come from.
Landon Gutierrez
15. Resistance at 30min 20SMA.
Adrian Butler
Nicholas Morris
Then I take my profit.
Daniel Perry
If you look at this, and say everything I have shown was just completely random, I have no idea what could convince you otherwise. I am always building a larger journal, but that takes time. I don;t have years of data.
Logan Johnson
i have my own algo which earned me 0.38ETH yesterday
Nolan King
This. My dad works at Ph.D. and he said you're gay too.
Camden Wilson
Unless you show your equity curve vs. B&H over year+ for multiple assets, you're just larping
Thomas Roberts
You have to realize that I'm just getting started. There are many things I still have to learn.
But if anyone looks at a market, and says "it's always random", then I could not agree. It's not what observable reality shows.
Jonathan Long
Have you backtested this?
Angel Bailey
I'm still building a strategy, I am inexperienced. I don't know entirely how the Dow works. I don't not yet have a strategy to backtest.
There's also the discretionary Human apsect that I wouldn't even know how to incorporate into any sort of algorithmic trading strategy.
But, the 200SMA is extremely important, that much I know.